review
Insect resistance to Bt crops: lessons
from the first billion acres
npg
© 2013 Nature America, Inc. All rights reserved.
Bruce E Tabashnik1, Thierry Brévault2 & Yves Carrière1
Evolution of resistance in pests can reduce the effectiveness of insecticidal proteins from Bacillus thuringiensis (Bt) produced by
transgenic crops. We analyzed results of 77 studies from five continents reporting field monitoring data for resistance to Bt crops,
empirical evaluation of factors affecting resistance or both. Although most pest populations remained susceptible, reduced efficacy
of Bt crops caused by field-evolved resistance has been reported now for some populations of 5 of 13 major pest species examined,
compared with resistant populations of only one pest species in 2005. Field outcomes support theoretical predictions that factors
delaying resistance include recessive inheritance of resistance, low initial frequency of resistance alleles, abundant refuges of non-Bt
host plants and two-toxin Bt crops deployed separately from one-toxin Bt crops. The results imply that proactive evaluation of the
inheritance and initial frequency of resistance are useful for predicting the risk of resistance and improving strategies to sustain the
effectiveness of Bt crops.
Transgenic crops are one of the most widespread and controversial
applications of biotechnology1–4. To reduce reliance on insecticide
sprays, scientists have genetically engineered corn and cotton plants to
make insecticidal proteins encoded by genes from the common bacterium Bacillus thuringiensis (Bt)5. These Bt proteins kill some devastating insect pests, but cause little or no harm to most other organisms,
including people4,5. Benefits of Bt crops include reduced insecticide
use, pest suppression, conservation of beneficial natural enemies,
increased yield and higher farmer profits6–12. The area planted with
Bt crops worldwide increased from 1.1 million hectares in 1996 to
66 million hectares in 2011, with a cumulative total of more than 420 million hectares (>1 billion acres) (Fig. 1). Bt corn accounted for 67% of corn
planted in the United States during 2012 (http://www.ers.usda.gov/Data/
BiotechCrops/) and Bt cotton accounted for 79–95% of cotton planted in
Australia, China, India and the United States during 2010 to 2012 (Fig. 2).
The remarkable ability of insects to adapt to insecticides and other
control tactics supports the conclusion that evolution of resistance by
pests is the main threat to the continued success of Bt crops13–23. Many
previous reviews have addressed pest resistance to Bt crops13–23, including a 2011 mini-review emphasizing four successful cases of the highdose⁄refuge resistance management strategy in North America22 and our
2009 review of 17 cases involving 11 species of lepidopteran pests and
four Bt toxins (B.E.T., Van Renburg, J.B.J. & Y.C.)21. Several papers have
compared field outcomes for resistance to Bt crops with predictions from
theory, but the rigor of these previous comparisons has been limited by
small sample sizes for both the field outcomes and the factors predicted
to affect resistance19–22.
1Department
of Entomology, University of Arizona, Tucson, Arizona, USA.
de coopération Internationale en Recherche Agronomique pour le
Développement, UPR 102, Montpellier, France. Correspondence should be
addressed to B.E.T. ([email protected]).
2Centre
Received 24 October 2012; accepted 26 March; published online 10 June 2013;
doi:10.1038/nbt.2597
510
Here we summarize the theory for managing pest resistance to Bt
crops, outline new criteria for categorizing evidence of field-evolved
resistance, review the global status of resistance to Bt crops based on
current field monitoring data, and test the correspondence between
theoretical predictions and observed patterns of field-evolved resistance. The criteria for categorizing field-evolved resistance described
and applied here explicitly acknowledge that resistance is not ‘all or
none’, which facilitates objective classification of monitoring data and
may help to gauge management actions appropriately, depending on the
severity of resistance. Compared with previous reviews on this topic,
the field monitoring data analyzed here are more recent and represent
more cases (24 in all), as well as larger and more diverse sets of Bt toxins
(six toxins from four Cry families) and pest species (13 species from
two insect orders). Using data from 77 studies published as of 2012, we
report the first statistical analyses of the association between observed
global patterns of field-evolved resistance and predicted effects of two
key biological parameters: dominance of resistance and initial resistance
allele frequency. The results provide insights that can be used proactively
to improve resistance management.
Theory for managing pest resistance to Bt crops
The refuge strategy has been the primary approach used worldwide to
delay pest resistance to Bt crops and has been mandated in the United
States, Australia and elsewhere8,16,23. Despite implementation of some
resistance management practices for conventional insecticides, the mandates for the refuge strategy are part of an unprecedented proactive effort
to slow resistance to Bt crops that recognizes both their value and the
strong threat of resistance. The concept underlying the refuge strategy is
that most of the rare resistant pests surviving on Bt crops will mate with
the relatively abundant susceptible pests from nearby refuges of host
plants without Bt toxins24–27. If inheritance of resistance is recessive, the
progeny from such matings will die on Bt crops, substantially delaying
the evolution of resistance. This approach is sometimes called the ‘highdose refuge strategy’ because it works best if the dose of toxin for insects
volume 31 number 6 JUNE 2013 nature biotechnology
re v ie w
70
7
Bt crops
Resistant species
✽
5
40
4
30
3
20
2
10
1
0
0
Figure 1 Planting of Bt crops globally and field-evolved resistance. Planting
of Bt crops globally each year and cumulative number of insect species with
field-evolved resistance and reduced efficacy reported. Planting of Bt crops
increased from 1.1 million hectares (ha) in 1996 to 66 million ha in 2011
(ref. 2). Field-evolved resistance associated with reduced efficacy of Bt crops
has been reported for five major target pests (year first detected): H. zea
(2002), S. frugiperda (2006), B. fusca (2007), P. gossypiella (2008) and
D. v. virgifera (2009) (Tables 1 and 2). *, For 2011, the number of species
with resistant populations may be underestimated because reports of fieldevolved resistance typically are published 2 or more years after resistance is
first detected.
Bt cotton (%)
eating Bt plants is high enough to kill all (or almost all) of the offspring
from matings between resistant and susceptible insects16 (B.E.T., Y.C. et
al.)15,27. Therefore, in theory, three key factors favor success of the refuge
strategy: first, recessive inheritance of resistance; second, low resistance
allele frequency; and third, abundant refuges of non-Bt host plants near
Bt crops16 (B.E.T., Y.C. et al.)15,21.
Two additional factors predicted to delay
100
100
Australia
resistance are fitness costs and incomplete
90
90 China
16
19,28,29
Cry1Ac
Cry1Ac
resistance (B.E.T., Y.C. et al.)
. Fitness
80
80
Cry1Ac + Cry2Ab
70
costs occur when fitness on non-Bt host plants
70
60
60
is lower for resistant insects than susceptible
50
50
insects, so that refuges select against resis40
40
tance28,29. Incomplete resistance occurs when
30
30
resistant insects can complete development on
20
20
Bt plants, but they are at a disadvantage com10
10
pared with resistant insects that develop on
0
0
100
100
non-Bt plants19,28.
India
90 United States
90
The dominance of resistance on a Bt crop
Cry1Ac
Cry1Ac
80
80
plant can be measured in terms of the paramCry1Ac + Cry2Ab
Cry1Ac + Cry2Ab or Cry1F
70
70
eter h, which varies from 0 for completely
60
60
16
recessive to 1 for completely dominant (Liu,
50
50
Y.B. & B.E.T)30. When such direct data are not
40
40
available, dominance can be assessed indirectly
30
30
20
20
by measuring survival of susceptible insects
10
10
on Bt plants31. This indirect assessment relies
0
0
on the idea that if Bt plants do not kill all or
nearly all homozygous susceptible insects,
Year planted
Year planted
they probably will not kill nearly all individuals heterozygous for resistance. If so, survival
is likely to be higher for the heterozygotes than Figure 2 Percentage of cotton hectares planted with Bt cotton producing one toxin (gray) or two toxins
(black) in four countries. All Bt cotton produced Cry1Ac. In Australia and India, all two-toxin cotton
for the homozygous susceptible insects, which produced Cry1Ac and Cry2Ab. In the United States from 2004 to 2012, 86% of two-toxin cotton
yields nonrecessive inheritance of resistance produced Cry1Ac and Cry2Ab and 14% produced Cry1Ac and Cry1F. The ranking of each country in terms
that accelerates adaptation16,27. Thus, the US of 2012 cotton production (percentage of world production) was 1 for China (27%), 2 for India (22%), 3
Environmental Protection Agency guidelines for the United States (15%) and 7 for Australia (3.4%) (see Supplementary Methods for details).
nature biotechnology volume 31 number 6 JUNE 2013
19
1996
1997
9
19 8
2099
0
20 0
20 01
0
20 2
0
20 3
0
20 4
2005
0
20 6
2007
0
20 8
0
20 9
1
20 0
2011
12
19
9
19 6
1997
9
19 8
2099
0
20 0
0
20 1
0
20 2
0
20 3
0
20 4
2005
0
20 6
2007
0
20 8
0
20 9
1
20 0
11
Bt cotton (%)
npg
© 2013 Nature America, Inc. All rights reserved.
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
Bt crops (million ha)
50
Resistant species
(reduced efficacy reported)
6
60
specify that high-dose Bt plants should kill at least 99.99% of susceptible
insects in the field31.
Mathematical modeling consistently predicts that resistance
will evolve more slowly if the initial resistance allele frequency is
low15,24,25,32,33. Although several papers22,34–38 propose (without theoretical or empirical evidence) that the success of the refuge strategy
requires an initial resistance allele frequency ≤0.001, modeling results
imply that the refuge strategy can be useful with much higher resistance allele frequencies, particularly if fitness costs are associated with
resistance24,28,33,39,40. For example, with recessive resistance and fitness
costs, refuges delayed resistance substantially in a model with an initial
resistance allele frequency of 0.3 (Y.C. & B.E.T.)28.
Refuge abundance can be measured for each pest in terms of the
percentage of its host plants that are non-Bt plants. If more than one
species of non-Bt host plant is available, the ‘effective’ refuge percentage
can be estimated by adjusting for the relative abundance of susceptible
pests produced on different host plant species41–45. The effective refuge
percentage can also be adjusted downward for the effects of treating
refuges with insecticides, because such treatments reduce the ability
of refuges to delay resistance41. Because pest movement and mating
patterns interact with the distribution and abundance of refuges and Bt
crop fields to affect evolution of resistance, spatially explicit approaches
are useful for assessing refuge effectiveness46.
Modeling results suggest that when inheritance of resistance is not
recessive, increasing refuge abundance can still substantially delay resistance. For example, results from a single-locus, two-allele model of a
generic pest with an initial resistance allele frequency of 0.001 suggest
that resistance can be delayed for >20 years with ≥5% refuges if resistance
is completely recessive (h = 0) and with >50% refuges if resistance is
partially dominant (h ≥ 0.4) (ref. 20).
First-generation Bt crops each produce a single Bt toxin, but many
second-generation Bt crops, named pyramids, produce two or more
511
re v ie w
Box 1 An alternative definition of resistance
npg
© 2013 Nature America, Inc. All rights reserved.
The Insecticide Resistance Action Committee (IRAC), composed
of members from more than a dozen major agrochemical and
biotech companies, aims to promote resistance management
strategies for insecticides and Bt crops to support sustainable
agriculture and improve public health (http://www.irac-online.
org/about/irac/). IRAC defines resistance as “a heritable change
in the sensitivity of a pest population that is reflected in the
repeated failure of a product to achieve the expected level of
control when used according to the label recommendation for
that pest species” (http://www.irac-online.org/about/resistance/).
The first part of the IRAC definition, “a heritable change in the
sensitivity of a pest population” and the definition of field-evolved
resistance (see main text) both emphasize a genetically based
decrease in susceptibility. The remainder of the IRAC definition
sets additional conditions that are problematic for identifying
distinct Bt toxins that are active against the same pest21,47. The assumption underlying this approach (which is not always true) is that selection
for resistance to one toxin does not cause cross-resistance to the other
toxins in the pyramid, so that insects resistant to all toxins in the pyramid
are extremely rare47,48. Other factors favoring success of pyramids match
those listed above for the refuge strategy, including abundant refuges
and the following conditions for each toxin in the pyramid: recessive
inheritance of resistance, low initial resistance allele frequency, fitness
costs and incomplete resistance21,39,47,48. Modeling results and smallscale experiments with noncommercial Bt broccoli plants indicate that
resistance to pyramids evolves faster if one-toxin plants are grown concurrently with two-toxin plants47. This occurs because the one-toxin
plants select for resistance to each toxin separately, which reduces the
advantage of the two-toxin plants47.
Field-evolved resistance: criteria and categories
Field-evolved (or field-selected) resistance is a genetically based decrease
in susceptibility of a population to a toxin caused by exposure of the
population to the toxin in the field14,21,49. A Web of Science search with
‘topic = field-evolved resistance’ identified 54 publications, starting with
two 1996 papers about resistance to Bt toxins produced by two independent research teams50,51 and including 31 papers published from 2010 to
2012. These 54 publications were authored by >150 academic, government and industry scientists from five continents and have been cited
>900 times, including 300 citations in 2012. Despite this widespread and
increasing use of the term ‘field-evolved resistance’, some scientists favor
alternative terms and definitions for resistance (Box 1).
Natural genetic variation affecting responses to Bt toxins usually
occurs in insect populations, with some alleles conferring susceptibility
and others conferring resistance. Before an insect population is exposed
to a Bt toxin, alleles conferring resistance are typically rare16,19. Fieldevolved resistance occurs when exposure of a field population to a toxin
increases the frequency of alleles conferring resistance in subsequent
generations21. Thus, detecting resistance alleles without demonstrating
that their frequency has increased in field populations does not constitute evidence of field-evolved resistance21.
The primary goal of monitoring resistance to Bt crops is to detect
field-evolved resistance early enough to enable proactive management
before field failures occur21. Resistance monitoring includes sampling
and testing of insects that survive on Bt crops as well as insects from
other sources, including non-Bt host plants. Failure to sample insects
from Bt crops favors underestimation of the frequency of resistance,
which can postpone detection of resistance21.
512
resistance objectively and for proactive detection and responses
to resistance. First, by the time a product has failed repeatedly,
it is usually too late to respond most effectively to resistance.
Second, the “expected level of control” is not specified, which
allows variation in interpretation, including changes over time in
expectations. Third, because the definition depends on the label
recommendation, resistance cannot occur in any species that is
not on the label, which excludes evolution of resistance in nontarget pests and non-pest species96,104. By contrast, the term
“field-evolved resistance” as defined here explicitly recognizes
that resistance results from evolution, enables objective
identification of resistance, facilitates proactive detection and
management of resistance, and applies to resistance in pest and
beneficial organisms. Various other definitions of resistance have
been proposed and discussed in depth elsewhere105,106.
Pest control problems associated with field-evolved resistance vary
from none to severe, depending on the frequency of resistant individuals, the extent to which resistance increases survival in the field, the
geographical distribution of resistant populations, the insect’s population density and the availability of alternative controls21. We define four
categories of field-evolved resistance: 1) >50% resistant individuals and
reduced efficacy of the Bt crop in the field has been reported; 2) >50%
resistant individuals and reduced efficacy is expected, but has not been
reported; 3) 1–6% resistant individuals; and 4) <1% resistant individuals. For categories 3 and 4, the percentage of resistant individuals is low
enough that reduced efficacy of the Bt crop in the field is not expected.
We adopt terms used previously and refer to cases in category 3 as an
‘early warning’ of resistance52 and cases in category 4 as ‘incipient resistance’53. In principle, an additional category could be 6–50% resistant
individuals, but none of the cases reviewed here was in that range. The
fifth category is cases in which monitoring data show no statistically
significant decrease in susceptibility.
Each case reviewed here involves evaluation of field-evolved resistance to one Bt toxin in populations of one pest species from one
country. Although initial detection of resistance can be based on data
from a single field population, all of the cases of field-evolved resistance reviewed here entail evidence of genetically based, decreased
susceptibility from several field populations. To classify each case
of field-evolved resistance into one of the four categories we define
above, we estimated the percentage of individuals resistant to a toxin
based on survival on a diet treated with a ‘diagnostic concentration’ of
the toxin that kills all or nearly all susceptible individuals, or survival
on intact Bt plants or parts of Bt plants containing that toxin. For
many cases, we also report the resistance ratio, which is the concentration of toxin killing 50% of insects tested (LC50) for a field-derived
strain divided by the LC50 for a conspecific susceptible strain. Large
increases in LC50 in field-selected populations yield resistance ratios
>10 and indicate that >50% of the population is resistant21.
Resistance monitoring data
Here we review 24 cases for which resistance monitoring data are published in peer-reviewed journals for 13 species of major lepidopteran
and coleopteran pests that are targeted by six Bt toxins in transgenic
corn and cotton in eight countries (Tables 1 and 2 and Supplementary
Tables 1–4).
Reduced efficacy reported or expected. The cumulative number of
major pest species with field-evolved resistance to Bt toxins in crops
volume 31 number 6 JUNE 2013 nature biotechnology
re v ie w
Table 1 Evaluation of field-evolved resistance in 24 cases involving 13 species of major pests targeted by Bt crops a
Pesta
Bt crop
Yearsb
High dosec
Low initial freq.d
South Africa
8
No
?e
Toxin
Country
>50% resistant individuals and reduced efficacy reported
B. fusca
Corn
Cry1Ab
D. v. virgifera
Corn
Cry3Bb
USA
7
No
No
H. zea
Cotton
Cry1Ac
USA
6
No
No
P. gossypiella
Cotton
Cry1Ac
India
6f
No
?
S. frugiperda
Corn
Cry1F
USA
3
No
?
Cotton
Cry2Ab
USA
2g
No
No
D. saccharalis
Corn
Cry1Ab
USA
10
No
No
H. armigera
Cotton
Cry1Ac
China
13
No
No
O. furnacalis
Corn
Cry1Ab
The Philippines
5
No
?
P. gossypiella
Cotton
Cry1Ac
China
13
No
?
H. armigera
Cotton
Cry1Ac
Australia
15
No
Yes
H. armigera
Cotton
Cry2Ab
Australia
8
Yes
No
H. punctigera
Cotton
Cry2Ab
Australia
8
Yes
No
D. grandiosella
Corn
Cry1Ab
USA
6
?
Yes
D. v. virgifera
Corn
Cry34/35Ab
USA
4
No
No
H. punctigera
Cotton
Cry1Ac
Australia
10
?
Yes
H. virescens
Cotton
Cry1Ac
USA
11
Yes
No
H. virescens
Cotton
Cry1Ac
Mexico
11
?
?
H. virescens
Cotton
Cry2Ab
USAfa
2
Yes
?
O. nubilalis
Corn
Cry1Ab
USA
15
No
Yes
O. nubilalis
Corn
Cry1Ab
Spain
4
?
?
P. gossypiella
Cotton
Cry1Ac
USA
13
Yes
No
P. gossypiella
Cotton
Cry2Ab
USA
5
Yes
Yes
S. nonagroides
Corn
Cry1Ab
Spain
7
?
Yes
>50% resistant individuals and reduced efficacy expected
H. zea
1–6% resistant individuals
npg
© 2013 Nature America, Inc. All rights reserved.
<1% resistant individuals
No decrease in susceptibility
aSee
Table 2 and Supplementary Tables 1–6 for details. D. v. virgifera is a coleopteran; the other 12 pests are lepidopterans. bYears elapsed between the first year of commercialization in the
region studied and: (i) for the six cases with >50% resistant individuals and reduced efficacy reported or expected (red and orange), the first year of field sampling that yielded evidence of resistance, or (ii) for all other cases, the most recent year of monitoring data reviewed here (see Table 2 and Supplementary Tables 1–4 for details). cBased on direct evaluation of recessive inheritance
of resistance for 12 cases with relevant data and on survival of susceptible individuals on Bt plants for 7 cases without such direct data (Supplementary Table 5). dBased on an initial resistance
allele frequency below the detection threshold; yes indicates initial screening did not detect any major resistance alleles (Supplementary Table 6). e”?” indicates answer could not be determined
with available data. fExcludes years when Bt cotton was grown illegally in India before it was commercialized in 2002 (refs. 81,91,92). Resistance was first detected in samples collected in
2008, 6 years after commercialization. If illegal planting started in 2000, the total years elapsed would be 8. gMay reflect some cross-resistance caused by selection with Cry1Ac21,93,94.
and reduced transgenic crop efficacy increased from one in 2005 to
five in 2010 (Figs. 1, 3 and 4). These five cases include resistance to
Bt corn in three pests (Busseola fusca, Diabrotica virgifera virgifera
and Spodoptera frugiperda) and resistance to Bt cotton in two pests
(Helicoverpa zea and Pectinophora gossypiella; Tables 1 and 2 and
Boxes 2 and 3). In each of these five cases, field-evolved resistance
Reduced efficacy
1–6% resistant
<1% resistant
Susceptible
2005
2010
Figure 3 Resistance of major pest species to Bt crops in 2005 and 2010.
For each pest species, the color indicates the status of the most resistant
population. In 2005, the only pest with resistant field populations was H.
zea; the other eight pests evaluated were susceptible. Data for 2005 (n =
9 species) are from reference 21. Data for 2010 (n = 13 species) are from
Table 1.
nature biotechnology volume 31 number 6 JUNE 2013
was detected fewer than 10 years after the Bt crop was commercialized. In a sixth case of field-evolved resistance of H. zea to Cry2Ab in
Bt cotton, the percentage of resistant individuals exceeded 50% for
several populations and reduced efficacy is expected, but has not yet
been reported (Box 4).
Early warning: 1–6% resistant individuals. In four cases, the percentage
of resistant individuals increased significantly to reach 1–6%, which is
not expected to reduce efficacy in the field (Table 1 and Supplementary
Table 2). For field populations exposed to Cry1Ac cotton in China for
13 years, maximum survival at a diagnostic concentration of Cry1Ac
was 2.6% for Helicoverpa armigera52 and 5.6% for P. gossypiella54, with
0% survival for susceptible control populations. For Ostrinia furnicalis
exposed to Cry1Ab corn in the Philippines, the maximum survival at
a diagnostic concentration of Cry1Ab increased 14-fold from 0.4% in
2007 to 5.5% in 2009 (ref. 55). For Diatraea saccharalis in Louisiana,
data from F2 screens show the frequency of alleles conferring resistance
to Cry1Ab corn increased eightfold from 0.0023 in 2004 to 0.018 in
2009 (ref. 56). We estimate that the percentage of resistant individuals in the populations sampled in Louisiana in 2009 was 1.0–2.4%,
based on the partial dominance of resistance (Supplementary Table
5) and an estimated frequency of heterozygous individuals of 0.031
(Supplementary Table 2).
513
re v ie w
Table 2 Bioassay data indicating field-evolved resistance to the toxins in Bt crops for five pests with >50% resistant individuals and
reduced efficacy reporteda
Bioassay data
Pest
Cry toxin
Country
Strains
Year comm.b testedc
Initial yeard
Final yeare
Control
valuee
Test valuef
References
surv.g
0.0%
64%
21,95
96
Parameter
B. fusca
1Ab
S. Africa
1998
2
2006
2006
Max.
B. fusca
1Ab
S. Africa
1998
8
2007
2007
Max. surv.h
0.0%
88%
H. zea
1Ac
USA
1996
2
2002
2002
Max. surv.i
0.0%
52%
97,98
H. zea
1Ac
USA
1996
64
1992
2004
Max. RRj
1.2
580
59,60
H. zea
1Ac
USA
1996
197
2002
2006
Max. RRj
40k
>1,000
59,97,99
P. gossypiella
1Ac
India
2002l
6
2007
2009
Survivalm
2.0%
72%
100
P. gossypiella
1Ac
India
2002l
2
2007
2009
Max. RRj
1.0
47
100
S. frugiperda
1F
USA
2003
8
1990
2008
Max. RRn
1.0
>356
75
S. frugiperda
1F
USA
2003
13
2010
2011
Max. RRo
1.0
>970
76
S. frugiperda
1F
USA
2003
13
2010
2011
Survivalp
0.8%
90%
76
D. v. virgifera
3Bb
USA
2003
9
2009
2009
Survivalq
17%r
52%s
101
D. v. virgifera
3Bb
USA
2003
13
2010
2010
Survivalq
5%t
74%u
102
npg
© 2013 Nature America, Inc. All rights reserved.
aCry1Ab,
Cry1F and Cry3Bb produced by Bt corn; Cry1Ac produced by Bt cotton. bFirst year Bt crop was grown commercially in the location monitored. cTotal number
of field-derived strains tested in bioassays. dInitial and final years during which field populations were sampled. eValue for parameter from one or more susceptible
strains, based on initial year unless noted otherwise. fValue for parameter from one or more field-selected resistant strains, based on final year unless noted otherwise.
gMaximum survival in the field at 18 days on Bt corn plants relative to non-Bt corn plants for a susceptible field population (control value) and a resistant field population (test value). hSurvival in the greenhouse at 35 days on Bt corn plants relative to non-Bt corn plants for the most susceptible field population (control value) and
the most resistant field population (test value). iSurvival at 4 days on Bt cotton leaves relative to non-Bt cotton leaves for a susceptible lab strain (control value) and
the most resistant field population (test value); in addition, four strains derived from the field in 2004 had >50% survival at a diagnostic concentration of Cry1Ac in
diet59. jMaximum resistance ratio, the highest LC50 for a field-derived strain divided by the LC50 for one or more susceptible strains. kThe maximum resistance ratio
of 40 in 2002 reflects field-evolved resistance that was detected in that year. lBt cotton was grown illegally in Gujarat, India, for at least 2 years before it was commercialized81,91,92. mMean survival at a concentration of 1 microgram Cry1Ac per ml diet in lab bioassays; resistance detected in a population from Amreli, Gujarat,
sampled in 2008. nField-selected resistant strains were derived from four populations sampled in Puerto Rico during 2007 to 2008 (test value); the control value is
based on a strain derived from a field population sampled from Georgia in 2005. The most resistant field populations had <50% mortality and growth inhibition at the
highest concentration tested, yielding a maximum resistance ratio >35 based on LC 50 values and >356 based on the concentration causing 50% growth inhibition
(IC50). oField-selected resistant strains were derived from four populations sampled in Puerto Rico during 2010 and 2011 (test value); the control value is the mean
for nine strains derived from US mainland populations sampled during 2010 and 2011. pMean survival at 7 days on Bt corn leaves relative to non-Bt corn leaves for
the strains derived in 2010 and 2011 from Puerto Rico (test value) and the US mainland (control value). qMean survival on Cry3Bb corn plants relative to non-Bt corn
plants in lab bioassays. rMean for five strains derived in 2010 from “control” fields in Iowa where severe corn rootworm damage was not seen. sMean for four strains
derived in 2009 from four “problem” fields in Iowa where growers reported severe corn rootworm injury to Bt corn fields planted with Cry3Bb corn (3 fields) or a combination of Cry3Bb corn and Cry34/35Ab corn (1 field). tMean for six control strains derived before Cry3Bb corn was commercialized (1995–2001) from fields in four
states. uMean for seven strains derived in 2010 from problem fields in Iowa.
Incipient resistance: <1% resistant individuals. For three cases that entail 2010–2011 (ref. 57). These results show that the statistically significant
H. armigera and Helicoverpa punctigera exposed to Bt cotton in Australia, yet small rises in resistance allele frequency characteristic of incipient
F1 and F2 screens detected statistically significant increases over time resistance do not necessarily indicate that further increases in resistance
in the frequency of alleles conferring resistance to Cry1Ac or Cry2Ab, are imminent.
yet the highest estimated percentage of resistant individuals was <1%
(Table 1 and Supplementary Table 3). Because of the low percentage of No decrease in susceptibility. Monitoring data provide strong evidence
resistant individuals after many years of extensive exposure to Bt cotton, of no decrease in susceptibility to toxins produced by Bt crops in 11
cases (Table 1 and Supplementary Table 4). These cases include evithese three cases exemplify successful resistance management.
Among these three cases, the maximum resistance allele frequency dence of no decreased susceptibility to each Bt toxin tested against all
detected is 0.048, based on results of the F1 screen for H. punctigera populations examined for four species: Diatraea grandiosella, Heliothis
resistance to Cry2Ab in Australia in 2008–2009 (refs. 53,57,58). Results virescens, Ostrinia nubilalis and Sesamia nonagrioides. In addition, the
from F1 screens conducted in 2008–2009 also
showed that the frequency of alleles conferring
resistance to Cry2Ab was eight times higher
in areas where Bt cotton was grown relative to
non-cropping areas (P < 0.0001, ref. 53). Based
on this difference between areas in the same
season and a significantly increased frequency
of resistance to Cry2Ab over time in Bt cotReduced efficacy
ton growing areas, Downes et al.53 termed this
>50% resistant
“incipient resistance.” Based on recessive inher1–6% resistant
<1% resistant
itance and Hardy-Weinberg equilibrium, they
Susceptible
2
estimated 0.2% (0.048 ) of H. punctigera larvae
were resistant to Cry2Ab in 2008–2009, which
is too low to reduce the efficacy of Bt cotton in Figure 4 Global status of field-evolved resistance to Bt crops. Each circle represents 1 of 24 cases
the field. Moreover, the frequency of resistance involving evaluation of field-evolved resistance to one toxin in Bt corn or Bt cotton in populations of one
to Cry2Ab did not increase from 2008–2009 to pest species from one country (Tables 1 and 2 and Supplementary Tables 1–4).
514
volume 31 number 6 JUNE 2013 nature biotechnology
re v ie w
Box 2 Field-evolved resistance to Bt corn with reduced efficacy reported
npg
© 2013 Nature America, Inc. All rights reserved.
Field-evolved resistance of S. frugiperda (fall armyworm) to Bt corn
producing Cry1F occurred in 3 years in the United States territory of
Puerto Rico75,76 (Tables 1 and 2). This is the fastest documented
case of field-evolved resistance to a Bt crop with reduced efficacy
reported and is consistent with worst-case scenarios envisioned in
1997 by some experts32,68. It is also the first case of resistance
leading to withdrawal of a Bt crop from the marketplace. High levels
of resistance persisted in Puerto Rico in 2011, 4 years after Cry1F
corn had been voluntarily withdrawn from sales76.
Field-evolved resistance to Bt corn producing Cry1Ab occurred
in B. fusca (maize stem borer) in South Africa in 8 years21,101
(Tables 1 and 2) and has some striking parallels with S. frugiperda
resistance to Cry1F corn. In both cases, proactive resistance
monitoring was not conducted and anecdotal evidence of reduced
data show no decreased susceptibility of D. v. virgifera to Cry34/35Ab
and of P. gossypiella to Cry1Ac and Cry2Ab in Arizona. Three cases in
the United States show no decrease in susceptibility after ≥10 years of
exposure to a toxin produced by a Bt crop: O. nubilalis to Cry1Ab in Bt
corn, and H. virescens and P. gossypiella to Cry1Ac in Bt cotton. Overall,
5 of the 24 cases show no decrease in susceptibility after ≥10 years of
exposure to the Bt crop; and 14 of the 24 cases (58%) show either no
decrease in susceptibility (11 cases) or <1% resistant individuals (3 cases)
after 2 to 15 years (mean = 9 years) (Table 1).
efficacy in the field preceded documentation of resistance with
bioassays73–76,101,102,107.
Bt corn producing Cry3Bb to kill beetles, particularly D. v. virgifera
(western corn rootworm), was first registered in the United States
in 2003 (ref. 108). By 2009, farmers planted Cry3Bb corn on 13
million ha, which was 36% of all corn in the United States100,109.
Field and laboratory data show that control problems in the field
during 2009 and 2010 were associated with resistance to Cry3Bb in
some Iowa populations of D. v. virgifera110–112. In ‘problem’ fields,
which had severe damage to Cry3Bb corn caused by rootworms,
Cry3Bb corn had been planted for 3–7 years110,111. A 2011 field
study of two of the problem fields identified in 2009 found that
D. v. virgifera emergence did not differ significantly between
Cry3Bb corn and non-Bt corn112.
Testing theory with data
The data from resistance monitoring studies reviewed here generally
confirm the main predictions from the evolutionary theory underlying the refuge and pyramid strategies for managing pest resistance to
Bt crops. As detailed below, resistance was less likely to evolve rapidly
if the high-dose standard was met (indicating recessive inheritance of
resistance), the initial resistance allele frequency was low, refuges were
abundant and Bt plants with two-toxin pyramids were grown separately
from one-toxin Bt plants.
Box 3 Field-evolved resistance to Bt cotton with reduced efficacy reported
Both cases of field-evolved resistance to Bt cotton with reduced
efficacy reported (Tables 1 and 2) have been controversial.
In India, Bt cotton hybrids generated by crossing a Bt cotton
cultivar with local non-Bt cotton cultivars were commercialized
in 2002, but illegal planting of Bt cotton hybrids began sooner
in the western state of Gujarat91,92. Resistance of P. gossypiella
(pink bollworm) to Bt cotton producing Cry1Ac (Bollgard) was
first detected with laboratory bioassays of the offspring of insects
collected from the field in 2008 in Gujarat97 (Table 2). Monsanto
(St. Louis) reported that their 2009 field monitoring confirmed
P. gossypiella resistance to Cry1Ac in four districts of Gujarat113.
This resistance seen in laboratory bioassays was associated with
unusually high abundance of larvae on Cry1Ac cotton and moths
caught in pheromone traps98,113,114.
A prominent Indian entomologist challenged the conclusion
of field-evolved resistance, claiming that resistance monitoring
should be based only on insects collected from non-Bt cotton,
yet Monsanto had collected larvae from Bt cotton plants98,114.
Monsanto aptly countered this criticism by stating that their
resistance monitoring based on insects collected from Bt
cotton in India is “standard practice”98. Indeed, testing insects
collected from Bt plants is an essential component of resistance
monitoring21,75,101,107,110–112,115.
Ironically, some of the dubious arguments disputing Monsanto’s
report of P. gossypiella resistance to Cry1Ac cotton in India
mirror those offered by Monsanto and others99 to challenge
documentation of H. zea (bollworm) resistance to Cry1Ac cotton
in the United States116. Researchers discovered the initial
evidence of field-evolved resistance of H. zea to Cry1Ac cotton
in the southeastern United States in 2002, 6 years after its
nature biotechnology volume 31 number 6 JUNE 2013
commercialization in that region59,117,118. The extensive evidence
confirming this case of resistance includes >50% survival at a
diagnostic concentration for four strains derived from the field in
2003 (refs. 59,115).
One of the primary arguments disputing the conclusion of fieldevolved resistance in this case was that “larval samples should
not be collected from Bt crops” for resistance monitoring99. As
noted above, testing insects sampled from Bt crops is critical
for monitoring resistance. Moreover, the evidence in this case
documents resistance in samples from non-Bt crops as well as from
Bt crops, including a strain derived from non-Bt cotton in 2004
that had a resistance ratio >500 (refs. 59,115,119). Another
challenge was that the evidence of field-evolved resistance came
entirely from the laboratory99. However, “unacceptable levels of
boll damage” observed in problem fields were associated with
decreased susceptibility to Cry1Ac in laboratory bioassays115,117,
similar to the evidence from India97,98,113.
In 2012, Luttrell and Jackson118 asserted that selection of
H. zea resistance to Cry1Ac in the laboratory before Bt cotton
was commercialized “argues against conclusions of field-evolved
resistance.” Yet, the selection experiment they cite60 demonstrates
that resistance alleles were present, but not common, before Bt
cotton was commercialized. This is reflected in the low LC50 of
Cry1Ac before laboratory selection for the strain derived from
the field in 1992, and the >100-fold increase in LC50 of this
strain caused by seven generations of selection with Cry1Ac60.
In the United States, the registration for Cry1Ac cotton expired
in September 2009 (ref. 21) and this product was replaced
progressively from 2003 to 2011 by cotton that produces two Bt
toxins, either Cry1Ac and Cry2Ab or Cry1Ac and Cry1F (Fig. 2).
515
re v ie w
npg
© 2013 Nature America, Inc. All rights reserved.
Box 4 Field-evolved resistance to Bt cotton with reduced efficacy likely
Like both cases of field-evolved resistance to Bt cotton producing
Cry1Ac (Box 3), the case of H. zea resistance to Cry2Ab in the
southeastern United States has been controversial. The initial
data documenting resistance in this case show a significant
increase in the proportion of populations screened that had an
LC50 value greater than the diagnostic concentration of toxin
(150 mg Cry2Ab per ml diet), which indicates >50% survival at
the diagnostic concentration21,93 (Supplementary Table 1). Based
on this criterion, the percentage of H. zea populations tested that
were resistant to Cry2Ab rose from 0% in 2002 to 50% in 2005,
only 2 years after commercialization of Bt cotton producing
Cry2Ab and Cry1Ac21,93. The percentage of populations with a
resistance ratio >10 also increased from 0% in 2002 to 50%
in 2005 (refs. 21,93). Three populations sampled from non-Bt
plants in Arkansas in 2005 had such low mortality in bioassays
that LC50 values could not be calculated, but were estimated to
be >400 mg Cry2Ab per ml diet93. The decreased susceptibility to
Cry2Ab detected in 2005, when cotton producing this toxin was
not common (Fig. 2), suggests that resistance to Cry1Ac caused
some cross-resistance to Cry2Ab93, which is consistent with data
showing a genetic correlation between resistance to these two
toxins94.
In addition, data from Arkansas show that mortality caused by a
diagnostic concentration of Cry2Ab decreased substantially in 2010
compared with the previous 4 years for field populations relative
to a susceptible laboratory strain120. This evidence of resistance
to Cry2Ab coincided with higher abundance of H. zea in the field
and increased insecticide sprays targeting H. zea on Bt cotton in
2010 (ref. 120). For the entire United States, the mean number of
insecticide sprays per hectare of Bt cotton directed primarily at
H. zea nearly doubled in 2009–2011 (0.88, s.e.m. = 0.1) compared
with 1999–2008 (0.48, s.e.m. = 0.03) (data from ref. 121; t-test,
t = 4.9, df = 11, P < 0.001). In the United States from 1999 to
2011, the percentage of Bt cotton producing two toxins increased
from 0% to 90% (Fig. 2), whereas the number of sprayings against
H. zea on Bt cotton tripled121.
In the five states of the midsouth region, sprays for H. zea per
hectare of Bt cotton were relatively low from 2004 to 2007 (mean
= 0.75, s.e.m. = 0.04), compared with 2000–2003 and 2008–
2010 (data from ref. 118; mean = 1.2, s.e.m. = 0.09; t = 3.9,
df = 9, P < 0.01). One explanation for this pattern is that fewer
sprays were needed during 2004 to 2007 because two-toxin plants
producing Cry1Ac and Cry2Ab initially had relatively high efficacy
against H. zea, but their efficacy declined because of resistance
to Cry2Ab. An alternative hypothesis is that sprays increased
because of increased planting of corn, which is a preferred host
for H. zea118. However, we found no association between the area
planted to corn and sprays for H. zea on Bt cotton in the midsouth
from 1999 to 2010 (data from ref. 118; r2 = 0.01, df = 10, P =
0.76). We also found no association between the area planted to
corn and sprays for H. zea on all cotton in Arkansas, Georgia and
Mississippi from 2000 to 2011 (Supplementary Tables 7–9 and
Supplementary Fig. 1).
Because some susceptible individuals can complete
development in the field on cotton producing Cry1Ac and
Cry2Ab122 and resistance to Cry1Ac in diet tests is associated
with increased survival on cotton leaves containing both
toxins, the increased survival of field-selected strains on diet
treated with diagnostic concentrations of Cry1Ac and Cry2Ab is
probably associated with increased survival in the field on cotton
plants producing both of these toxins 21,119. Although Luttrell
and Jackson118 state that they did not find strong evidence
of “sustained loss of field control or increased resistance
levels over time,” they conclude, “From a practical farm-level
perspective, effective control of bollworm [H. zea] requires
supplemental insecticides, even on dual-gene Bt cottons.”
High dose. Field outcomes show that resistance was less likely to evolve
quickly if plants met the high-dose standard indicating that resistance
was inherited as a recessive trait (Table 1 and Fig. 5). Available data
enabled evaluation of this factor for 19 cases, based on direct assessment
of dominance (h) (12 cases) or indirect assessment derived from survival
of susceptible pests on Bt plants in the field (7 cases) (Supplementary
Table 5). Bt plants met the high-dose standard in six of nine (67%) cases
with either no decrease in susceptibility or <1% resistant individuals, but
not in any of the ten cases with ≥1% resistant individuals (Fisher’s exact
test, P = 0.003; Table 1).
A compelling contrast confirming the importance of the high-dose
criterion is seen between the rapid evolution of resistance to Cry1Ac in
Bt cotton by H. zea, but not by the closely related pest H. virescens (Table
1). Cry1Ac cotton met the high-dose standard against H. virescens, but
not H. zea (Table 1). These two polyphagous pests that attack cotton
were sampled from the same region and tested side-by-side in some
studies59,60. Moreover, evolution of resistance to insecticides other than
Bt toxins has been faster in H. virescens than H. zea61,62, which refutes
the alternative hypothesis that resistance generally evolves faster in
H. zea than H. virescens.
One of the three exceptional cases in which the high-dose standard
was not met and the percentage of resistant individuals was <1% involves
D. v. virgifiera and Bt corn producing Cry34/35Ab (Table 1). In this case,
the monitoring data cover only 4 years since commercialization, during
which adoption of this product has been limited63.
The other two exceptions involve cases with <1% resistant individuals after 15 years of exposure to Bt crops: O. nubilalis and Cry1Ab corn
in the United States and H. armigera and Bt cotton in Australia (Table
1). In both cases, inheritance of resistance was completely recessive on
young plants (h = 0), but not on older plants (h = 0.31 for O. nubilalis
and 0.63 for H. armigera)64–66. These direct estimates of dominance
516
High dose
Not high dose
Figure 5 Resistance to Bt crops and dose criterion. Resistance evolved more
slowly when the high-dose criterion was met (left, n = 6 cases) than when
it was not met (right, n = 13 cases). Red, >50% resistance and reduced
efficacy reported; orange, >50% resistance and reduced efficacy expected;
yellow, 1–6% resistant individuals; blue, <1% resistant individuals;
green, no decrease in susceptibility (see Table 1 and text for details).
volume 31 number 6 JUNE 2013 nature biotechnology
re v ie w
npg
© 2013 Nature America, Inc. All rights reserved.
are based on survival of F1 progeny on Bt plants to the adult stage for
H. armigera65,66, but only for 15 days for O. nubilalis, which might
overestimate h for this pest64. In the experiments with H. armigera and
Bt cotton, the concentration of Cry1Ac was 75% lower in the old plants
compared with the young plants66.
In both of these cases, non-Bt crop refuges were abundant. For H.
armigera in Australia, the mean percentage of non-Bt cotton was 73%
from 1996 to 2003 (range, 40–90%) when Cry1Ac cotton was planted,
and 15% (range, 6–30%) from 2004 to 2011, when two-toxin Bt cotton
producing Cry1Ac and Cry2Ab replaced Cry1Ac cotton57 (Fig. 2). For
O. nubilalis and Bt corn in the United States, the minimum percentage
of corn planted with non-Bt corn for any state for a given year was 24%,
which occurred in Iowa in 2010 and 2012 (ref. 67). From 1996 to 2012,
the mean was 53% non-Bt corn in Iowa67. Overall, the results show that
rapid evolution of resistance is less likely when the high-dose standard
is met, and in some cases when this criterion is not met throughout the
growing season, resistance can be delayed for more than a decade with
abundant refuges.
Low initial resistance allele frequency. The monitoring data show that
rapid resistance evolution was less likely when the initial resistance allele
frequency was low (Table 1). The initial resistance allele frequency was
below the detection threshold (no major resistance alleles detected, estimated frequency = 0) in 6 of 11 cases (55%) with either no decrease in
susceptibility or <1% resistant individuals, compared with 0 of 5 cases
with >1% resistant individuals (Fisher’s exact test, one-tailed P = 0.058;
Table 1 and Supplementary Table 6).
With a criterion of an initial resistance allele frequency <0.001,
however, the association with resistance was not significant. The initial resistance allele frequency was <0.001 in 7 of 11 cases (64%) with
no decrease in susceptibility or <1% resistant individuals versus 2 of
5 cases (40%) with ≥1% resistant individuals (Fisher’s exact test, onetailed P = 0.37). Moreover, in two of the three cases from the United
States with no decrease in susceptibility for more than a decade, the
estimated initial resistance allele frequency was not <0.001; instead it
was 0.0015 for H. virescens in four southern states and 0.16 for P. gossypiella in Arizona68–71 (Table 1 and Supplementary Tables 4 and 6). In
laboratory-selected strains of these pests, resistance to Cry1Ac is recessive68,71 (Supplementary Table 5). In addition, refuges were abundant
for the first decade in both of these cases. The mean statewide percentage of cotton planted to non-Bt cotton from 1996 to 2005 was 42% in
Arizona9 and 50% in Arkansas, which had one of the highest adoption
rates of Bt cotton of any state where H. virescens was monitored59,69,72.
These results support the prediction from modeling studies that even
when the initial resistance allele frequency exceeds 0.001, resistance can
be delayed substantially, particularly if inheritance of resistance is recessive and refuges are abundant28. For P. gossypiella in Arizona, substantial
fitness costs and incomplete resistance probably also helped to delay
resistance28,71.
Refuges. Consistent with previous reviews based on relatively limited
data20–22, the more extensive monitoring data reviewed here support
the prediction that abundant refuges can delay resistance. Results from
grower surveys in South Africa imply that the low abundance of refuges of
non-Bt corn hastened evolution of B. fusca resistance to Bt corn producing
Cry1Ab73,74. On average, from 1998 to 2004, fewer than 30% of the farmers planting Bt corn in the Vaalharts area of South Africa complied with
contracts requiring them to plant non-Bt corn refuges73. In addition, precommercialization field data showing 2– 3% survival of susceptible larvae
on Cry1Ab corn relative to non-Bt corn indicate that this Bt corn does
not meet the high-dose standard against B. fusca (Supplementary Table
nature biotechnology volume 31 number 6 JUNE 2013
5). Thus, available evidence suggests that low refuge abundance and nonrecessive inheritance of resistance accelerated evolution of resistance by
B. fusca to Bt corn.
As with B. fusca, low refuge abundance and failure to meet the highdose criterion apparently accelerated evolution of S. frugiperda resistance
to Cry1F corn in Puerto Rico75,76. Based on mortality at the highest
concentration of Cry1F tested, resistance was partially recessive76
(h = 0.14), which does not meet the high-dose standard of at least 95%
mortality of heterozygotes31 (h ≤ 0.05). Although the levels of Cry1F in
Bt corn are “close to high dose” against this pest76, modeling results suggest that such moderate doses can cause faster resistance evolution than
either higher doses that kill all or nearly all heterozygotes, or lower doses
that allow substantial survival of susceptibles16 (B.E.T. & Croft, B.A.)25.
Cross-resistance to Cry1F caused by exposure to Cry1A toxins in sprays
and Bt corn might have also promoted resistance to Cry1F-producing
corn76. Evolution of resistance in this case was probably also accelerated
by continuous exposure to Bt corn during as many as 10 generations per
year, which translates to 30 generations of selection in 3 years76.
A scarcity of refuges in India and China may have promoted faster
evolution of P. gossypiella resistance to Cry1Ac cotton in these two countries compared with the United States (Table 1), where refuges have been
relatively abundant and high compliance with the refuge strategy was
documented by our team in Arizona (Y.C., B.E.T et al.)77,78. Regulations
in India mandate refuges of non-Bt cotton, but apparently compliance
has been low79,80. China has not required non-Bt cotton refuges, and the
non-Bt cotton percentage decreased to 8% in 2008 and 6% in 2009 and
2010 in six provinces of the Yangtze River Valley54.
Another factor accelerating P. gossypiella resistance in India and China
might be a lower concentration of Cry1Ac in the types of Bt cotton
grown there compared with the varieties grown in the United States.
In side-by-side field trials conducted in China, the abundance of P. gossypiella larvae was about five times higher on the predominant Bt cotton
variety grown in China compared with a Bt cotton variety grown on a
limited basis in the United States54. Although the efficacy and toxin
concentrations have not been compared directly among the most popular types of Bt cotton grown in these countries, survival of susceptible
P. gossypiella in the field was higher in both India and China than the
United States, and the high-dose standard was met in the United States,
but not in the two Asian countries (Table 1 and Supplementary Table 5).
It is unclear why P. gossypiella resistance to Cry1Ac is a much more serious problem in India than in China (Tables 1 and 2 and Supplementary
Table 2). However, unlike the true-breeding varieties of Bt cotton planted
in China, the United States, and elsewhere, hybrids account for nearly
all Bt cotton planted in India80. In 2009, >500 Bt cotton hybrids were
approved for planting in India80. Some of these diverse Bt cotton hybrids
and the unapproved Bt cotton grown in India81 may have lower toxin
concentrations than the Bt cotton varieties grown in China.
Comparing field outcomes and refuge abundance in Australia, China
and the United States for three congeneric pests (H. armigera, H. punctigera and H. zea) provides useful lessons for managing resistance when
the high-dose standard is not met (Table 1 and Fig. 2). Cotton plants
producing Cry1Ac or both Cry1Ac and Cry2Ab do not meet the highdose standard for any of these three pests (Table 1 and Supplementary
Table 5). After more than a decade of exposure to Bt cotton, the frequency of resistant individuals remained <1% for H. armigera and
H. punctigera in Australia for Cry1Ac and Cry2Ab, increased to between
1% and 5% for H. armigera in China for Cry1Ac, and exceeded 50% for
some populations of H. zea in the southeastern United States for both
Cry1Ac and Cry2Ab (Table 1).
Of the three countries, Australia has applied the most stringent refuge
requirements, which may have substantially delayed resistance. For cotton
517
re v ie w
npg
© 2013 Nature America, Inc. All rights reserved.
producing only Cry1Ac, the minimum percentage of non-Bt cotton
required on each farm in Australia was 70% from 1996 to 2003 (ref.
57) versus 4% in the United States26,82. For two-toxin cotton, Australia
requires 10% non-Bt cotton or the equivalent in terms of other non-Bt
crop host plants on each farm83, whereas the United States has eliminated refuge requirements in most regions84.
In China, however, virtually all Bt cotton planted produces only
Cry1Ac (Fig. 2) and refuges of non-Bt cotton have not been required52.
Nonetheless, non-Bt host plants other than cotton accounted for >92%
of the cropping area planted to H. armigera host plants from 1997–2006
(ref. 7), which probably slowed resistance. Although H. zea in the United
States also uses non-Bt host plants other than cotton, one of its major
alternative hosts is Bt corn producing Cry1Ab, which is expected to
select for cross-resistance to Cry1Ac41,43. Taking this and other factors
including insecticide sprays into account, the ‘effective refuge’ for H. zea
during the three generations in which it feeds on cotton was meticulously estimated as 39% in Arkansas for 2001–2005 (ref. 41).
Pyramids. Field outcomes are consistent with the prediction that resistance to pyramids will evolve faster if two-toxin plants are grown at the
same time as plants producing only one of the toxins in the pyramid47. In
the United States, farmers planted one-toxin cotton producing Cry1Ac
concurrently with two-toxin cotton producing Cry1Ac and Cry2Ab
from 2004 to 2010, whereas Australian growers completely replaced
Cry1Ac cotton with two-toxin cotton during 2004 (Fig. 2). As noted
above, the frequency of resistance to both toxins has exceeded 50% for
some populations of H. zea in the United States, whereas it has remained
<1% for H. armigera and H. punctigera in Australia (Table 1).
In principle, faster evolution of resistance in H. zea than in H. armigera could also reflect higher initial resistance allele frequencies or more
dominant inheritance of resistance in H. zea21. The available data suggest
that initial resistance allele frequencies for Cry1Ac and Cry2Ab were
not significantly higher for H. zea than for H. armigera (Supplementary
Table 6), but resistance to both toxins appears to be more dominant in
H. zea (Supplementary Table 5).
Conclusions
From 2005 to 2010, the data available to assess the effectiveness of
resistance management tactics for Bt crops increased dramatically and
the number of major target pests with some populations resistant to Bt
crops and reduced efficacy reported surged from one to five (Tables 1
and 2 and Figs. 1, 3 and 4). The increase in documented cases of resistance likely reflects increases in the area planted to Bt crops (Fig. 1), the
cumulative duration of pest exposure to Bt crops, the number of pest
populations exposed and improved monitoring efforts.
Our review of field-evolved resistance to Bt crops based on monitoring data for up to two decades from 24 cases in eight countries generally
confirms the principles of resistance management based on evolutionary
theory. As predicted, factors associated with sustained susceptibility to
the Bt toxins in transgenic crops are a toxin concentration that meets the
high-dose standard and thus renders inheritance of resistance recessive
(see Theory section above), a low initial frequency of resistance alleles,
and abundant refuges of non-Bt host plants near Bt crops that promote
survival of susceptible insects.
Before commercialization, scientists can evaluate insect responses
to Bt crops to determine if the high-dose standard is met and if the
initial frequency of resistance is low, using the techniques described
in the studies reviewed here (Supplementary Tables 5 and 6). Because
resistant strains are often not available before commercialization, the
high-dose standard can be assessed proactively by measuring survival of
susceptible insects on Bt crops31. In parallel, estimates of the frequency
of individuals with a genetically based decrease in susceptibility relative
to conspecific individuals can be made proactively with bioassays of
field-derived strains using Bt plants, Bt plant parts, or diagnostic concentrations of toxin in diet. Although F2 screens have been especially
useful for detecting rare recessive resistance alleles, the modeling and
empirical results reviewed here do not support the idea that it is critical
to determine if the initial resistance allele frequency is <0.001.
The relevant theory and data suggest that if the criteria for high dose
and low initial frequency are met, resistance can be delayed with limited
refuges. Conversely, if these criteria are not met, resistance is likely to
evolve rapidly unless refuges are abundant. Therefore, systematic assessment of these criteria can be used proactively to enhance resistance
management. Moreover, if reporting the assessment of these criteria
becomes standard practice, the data available for testing predictions will
increase steadily, thereby facilitating refinements in resistance management strategies.
In the past decade, farmers in the United States, India and Australia
have shifted largely from planting first-generation transgenic plants producing one Bt toxin to using ‘pyramids’ that produce two or more distinct Bt toxins active against a particular pest (Fig. 2). The limited field
data available for pyramids confirm predictions from theory and smallscale experiments with a model system indicating that pyramids work
best when implemented proactively47, as has been done in Australia57.
Conversely, when a pyramid of two toxins is adopted after resistance is
no longer rare to one of the toxins, the benefits of this approach seem to
Table 3 Bt toxin pyramids used proactively and separately from one-toxin plants or remedially and concurrent with one-toxin plants
Pest
Crop
Country
Toxins in pyramid21,57,63
Resistance detecteda
Proactive and separate from one-toxin plants
H. armigera
Cotton
Australia
Cry1Ac, Cry2Ab
None
H. punctigera
Cotton
Australia
Cry1Ac, Cry2Ab
None
Remedial and concurrent with one-toxin plants
D. v. virgifera
Corn
USA
Cry3Bb, Cry34/35Ab
Cry3Bb
H. zea
Cotton
USA
Cry1Ac, Cry2Ab
Cry1Ac
H. zea
Cotton
USA
Cry1Ac, Cry1F
Cry1Ac
P. gossypiella
Cotton
India
Cry1Ac, Cry2Ab
Cry1Ac
S. frugiperda
Corn
USA
Cry1F, Cry1A.105b, Cry2Ab
Cry1F
aResistance detected to one of the toxins in a pyramid before the pyramid completely replaced single-toxin Bt crops producing one of the toxins in the pyramid.
Monitoring data and references are provided in Supplementary Tables 3 and 4 for H. armigera and H. punctigera, and in Table 2 for the four other pests. bCry1A.105
is a chimeric Bt toxin with its amino acid sequence 99% identical to Cry1F for domain III, identical to Cry1Ab for domain I, and identical to Cry1Ac for domain II
and C terminus95. Although data evaluating S. frugiperda responses to Cry1A.105 have not been reported, cross-resistance to Cry1A.105 is expected in Puerto Rico
because populations there have been selected for resistance to each of its three parent toxins: Cry1F in Bt corn, and Cry1Ab and Cry1Ac in sprays 75,76. For S. frugiperda populations and families from Puerto Rico resistant to Cry1F, resistance ratios for Cry1Ab and Cry1Ac ranged from 12 to 89 (refs. 75,103).
518
volume 31 number 6 JUNE 2013 nature biotechnology
npg
© 2013 Nature America, Inc. All rights reserved.
re v ie w
be greatly reduced, as exemplified by resistance to Cry2Ab in Bt cotton
for H. zea in the United States (Table 1). In several other cases, pyramids
are also being used as a remedial tactic following documented resistance
to Bt crops producing only one of the toxins in the pyramid (Table 3).
Although all of the data reviewed here involve crystalline (Cry) Bt
toxins, transgenic crops producing vegetative insecticidal proteins (Vips)
from Bt are registered in the United States21 and their use is expected
in Australia85. In addition, genetically engineered Cry toxins that were
more effective than native Bt toxins against resistant strains of several
pests in laboratory tests might eventually broaden options for managing insects with transgenic plants18 (B.E.T. et al.)86. Scientists are also
developing transgenic plants that control insects by means of RNA interference and fusion proteins87–90.
Even with a wider range of approaches used to engineer plants for
protection against insects, resistance management will continue to be
essential. Based on the 24 cases reviewed here, pests can evolve resistance
to toxins in Bt crops in as few as 2 years under the worst circumstances;
under the best circumstances, however, efficacy can be sustained for
15 years or more. Although regulations in the United States and elsewhere mandate refuges of non-Bt host plants for some Bt crops, farmer
compliance is not uniformly high and the required refuge percentages
may not always be large enough to achieve the desired delays in evolution of resistance3,63,73–76,79–81. Both in theory and practice, using Bt
crops in combination with other tactics as part of integrated pest management may be especially effective for delaying pest resistance9. We
hope that the lessons learned from the first billion acres of Bt crops will
improve resistance management strategies in the future.
Note: Supplementary information is available in the online version of the paper.
ACKNOWLEDGMENTS
We thank A. Yelich for assistance with figures, and D. Crowder, L. Masson and
M. Sisterson for providing comments. This work was supported by US Department
of Agriculture (USDA) Agriculture and Food Research Initiative Grant 2008-353020390 and USDA Biotechnology Risk Assessment Grant 2011-33522-30729.
Author Contributions
Y.C., T.B. and B.E.T. analyzed data. B.E.T. wrote the paper. All authors discussed the
results and commented on the manuscript.
COMPETING FINANCIAL INTERESTS
The authors declare competing financial interests: details are available in the online
version of the paper.
Reprints and permissions information is available online at http://www.nature.com/
reprints/index.html.
1.
National Research Council. The Impact of Genetically Engineered Crops on Farm
Sustainability in the United States (National Academies Press, Washington, DC,
2010).
2. James, C. Global status of commercialized biotech/GM crops: 2011. ISAAA Briefs
43 (ISAAA, Ithaca, NY, 2011).
3. Alyokhin, A. Scant evidence supports EPA’s pyramided Bt corn refuge size of 5%.
Nat. Biotechnol. 29, 577–578 (2011).
4. Mendelsohn, M., Kough, J., Vaituzis, Z. & Matthews, K. Are Bt crops safe? Nat.
Biotechnol. 21, 1003–1009 (2003).
5. Sanahuja, G., Banakar, R., Twyman, R., Capell, T. & Christou, P. Bacillus thuringiensis: a century of research, development and commercial applications. Plant
Biotechnol. J. 9, 283–300 (2011).
6.Wu, K.M., Lu, Y.H., Feng, H.Q., Jiang, Y.Y. & Zhao, J.Z. Suppression of cotton bollworm in multiple crops in China in areas with Bt toxin-containing cotton. Science
321, 1676–1678 (2008).
7. Carpenter, J.E. Peer-reviewed surveys indicate positive impact of commercialized
GM crops. Nat. Biotechnol. 28, 319–321 (2010).
8. Hutchison, W.D. et al. Areawide suppression of European corn borer with Bt maize
reaps savings to non-Bt maize growers. Science 330, 222–225 (2010).
9. Tabashnik, B.E. et al. Suppressing resistance to Bt cotton with sterile insect releases.
Nat. Biotechnol. 28, 1304–1307 (2010).
10. Edgerton, M.D. et al. Transgenic insect resistance traits increase corn yield and yield
stability. Nat. Biotechnol. 30, 493–496 (2012).
11. Kathage, J.K. & Qaim, M. Economic impacts and impact dynamics of Bt (Bacillus
thuringiensis) cotton in India. Proc. Natl. Acad. Sci. USA 109, 11652–11656
nature biotechnology volume 31 number 6 JUNE 2013
(2012).
12. Lu, Y., Wu, K., Jiang, Y., Guo, Y. & Desneux, N. Widespread adoption of Bt cotton and
insecticide decrease promotes biocontrol services. Nature 487, 362–365 (2012).
13. Onstad, D. Insect Resistance Management: Biology, Economics, and Prediction
(Academic Press, London, 2008).
14. Heckel, D.G. Insecticide resistance after Silent Spring. Science 337, 1612–1614
(2012).
15. Tabashnik, B.E. Evolution of resistance to Bacillus thuringiensis. Annu. Rev. Entomol.
39, 47–79 (1994).
16. Gould, F. Sustainability of transgenic insecticidal cultivars: integrating pest genetics
and ecology. Annu. Rev. Entomol. 43, 701–726 (1998).
17. Ferré, J. & Van Rie, J. Biochemistry and genetics of insect resistance to Bacillus
thuringiensis. Annu. Rev. Entomol. 47, 501–533 (2002).
18. Pardo-López, L., Bravo, A. & Soberón, M. Bacillus thuringiensis insecticidal threedomain toxins: mode of action, insect resistance and consequences for crop protection. FEMS Microbiol. Rev. 37, 3–22 (2013).
19. Carrière, Y., Crowder, D.W. & Tabashnik, B.E. Evolutionary ecology of insect adaptation to Bt crops. Evol. Appl. 3, 561–573 (2010).
20. Tabashnik, B.E., Gassmann, A.J., Crowder, D.W. & Carrière, Y. Insect resistance to
Bt crops: evidence versus theory. Nat. Biotechnol. 26, 199–202 (2008).
21. Tabashnik, B.E., Van Rensburg, J.B.J. & Carrière, Y. Field-evolved insect resistance to
Bt crops: definition, theory, and data. J. Econ. Entomol. 102, 2011–2025 (2009).
22. Huang, F., Andow, D.A. & Buschman, L. Success of the high-dose/refuge resistance
management strategy after 15 years of Bt crop use in North America. Entomol. Exp.
Appl. 140, 1–16 (2011).
23. Tabashnik, B.E. Delaying insect resistance to transgenic crops. Proc. Natl. Acad. Sci.
USA 105, 19029–19030 (2008).
24. Georghiou, G.P. & Taylor, C.E. Genetic and biological influences in the evolution of
insecticide resistance. J. Econ. Entomol. 70, 319–323 (1977).
25. Tabashnik, B.E. & Croft, B.A. Managing pesticide resistance in crop-arthropod complexes: interactions between biological and operational factors. Environ. Entomol.
11, 1137–1144 (1982).
26. US Environmental Protection Agency. The Environmental Protection Agency’s White
Paper on Bt Plant-pesticide Resistance Management <http://www.epa.gov/EPAPEST/1998/January/Day-14/paper.pdf> (EPA, 1998).
27. Tabashnik, B.E., Gould, F. & Carrière, Y. Delaying evolution of insect resistance
to transgenic crops by decreasing dominance and heritability. J. Evol. Biol. 17,
904–912 (2004).
28. Carrière, Y. & Tabashnik, B.E. Reversing insect adaptation to transgenic insecticidal
plants. Proc. Biol. Sci. 268, 1475–1480 (2001).
29. Gassmann, A.J., Carrière, Y. & Tabashnik, B.E. Fitness costs of insect resistance to
Bacillus thuringiensis. Annu. Rev. Entomol. 54, 147–163 (2009).
30. Liu, Y.B. & Tabashnik, B.E. Inheritance of resistance to Bacillus thuringiensis toxin
Cry1C in the diamondback moth. Appl. Environ. Microbiol. 63, 2218–2223 (1997).
31. US Environmental Protection Agency. Final Report of the Subpanel on Bacillus
thuringiensis (Bt) Plant-Pesticides and Resistance Management, February, 1998
<http://www.epa.gov/scipoly/sap/meetings/1998/0298_mtg.htm> (EPA, 1998).
32. Roush, R.T. Bt-transgenic crops: just another pretty insecticide or a chance for a new
start in resistance management? Pestic. Sci. 51, 328–334 (1997).
33. Pittendrigh, B.R. et al. “Active” refuges can inhibit the evolution of resistance in
insects towards transgenic insect-resistant plants. J. Theor. Biol. 231, 461–474
(2004).
34. Roush, R.T. Managing pests and their resistance to Bacillus thuringiensis: can transgenic crops be better than sprays? Biocontrol Sci. Technol. 4, 501–516 (1994).
35. Andow, D.A., Olson, D.M., Hellmich, R.L., Alstad, D.N. & Hutchison, W.D. Frequency
of resistance to Bacillus thuringiensis toxin Cry1Ab in an Iowa population of European
corn borer (Lepidoptera: Crambidae). J. Econ. Entomol. 93, 26–30 (2000).
36. Bourguet, D. et al. Frequency of alleles conferring resistance to Bt maize in French
and US corn belt populations of the European corn borer, Ostrinia nubilalis. Theor.
Appl. Genet. 106, 1225–1233 (2003).
37. Stodola, T.J. et al. Frequency of resistance to Bacillus thuringiensis toxin Cry1Ab in
southern United States corn belt populations of European corn borer (Lepidoptera:
Crambidae). J. Econ. Entomol. 99, 502–507 (2006).
38. Huang, F., Parker, R., Leonard, R. & Yong, Y. & Liu, J. Frequency of resistance alleles
to Bacillus thuringiensis-corn in Texas populations of sugarcane borer, Diatraea saccharalis (F.) (Lepidoptera: Crambidae). Crop Prot. 28, 174–180 (2009).
39. Gould, F., Cohen, M.B., Bentur, J.S., Kennedy, G.G. & Van Duyn, J. Impact of small
fitness costs on pest adaptation to crop varieties with multiple toxins: a heuristic
model. J. Econ. Entomol. 99, 2091–2099 (2006).
40. Pan, Z. et al. Western corn rootworm (Coleoptera: Chrysomelidae) dispersal and
adaptation to single-toxin transgenic corn deployed with block or blended refuge.
Environ. Entomol. 40, 964–978 (2011).
41. Gustafson, D.I., Head, G.P. & Caprio, M.A. Modeling the impact of alternative hosts
on Helicoverpa zea adaptation to Bollgard cotton. J. Econ. Entomol. 99, 2116–2124
(2006).
42. Baker, G.H., Tann, C.T. & Fitt, G.P. Production of Helicoverpa spp. (Lepidoptera,
Noctuidae) from different refuge crops to accompany transgenic cotton plantings in
eastern Australia. Aust. J. Agric. Res. 59, 723–732 (2008).
43. Head, G. et al. Spatial and temporal variability in host use by Helicoverpa zea as
measured by analyses of stable carbon isotope ratios and gossypol residues. J. Appl.
Ecol. 47, 583–592 (2010).
44. O’Rourke, M.E., Sappington, T.W. & Fleischer, S.J. Managing resistance to Bt crops in
a genetically variable insect herbivore, Ostrinia nubilalis. Ecol. Appl. 20, 1228–1236
519
npg
© 2013 Nature America, Inc. All rights reserved.
re v ie w
(2010).
45. Brévault, T., Nibouche, S., Achaleke, J. & Carrière, Y. Assessing the role of non-cotton
refuges in delaying Helicoverpa armigera resistance to Bt cotton in West Africa. Evol.
Appl. 5, 53–65 (2012).
46. Carrière, Y. et al. Large-scale, spatially explicit test of the refuge strategy for delaying
insecticide resistance. Proc. Natl. Acad. Sci. USA 109, 775–780 (2012).
47. Zhao, J.-Z. et al. Concurrent use of transgenic plants expressing a single and two
Bacillus thuringiensis genes speeds insect adaptation to pyramided plants. Proc.
Natl. Acad. Sci. USA 102, 8426–8430 (2005).
48. Brévault, T. et al. Potential shortfall of pyramided transgenic cotton for insect resistance management. Proc. Natl. Acad. Sci. USA 110, 5806--5811 (2013).
49. National Research Council. Pesticide Resistance: Strategies and Tactics for
Management (National Academy Press, Washington D.C., 1986).
50. Liu, Y.B., Tabashnik, B.E. & Pusztai-Carey, M. Field-evolved resistance to Bacillus
thuringiensis toxin CryIC in diamondback moth (Lepidoptera: Plutellidae). J. Econ.
Entomol. 80, 798–804 (1996).
51. Tang, J.D. et al. Toxicity of Bacillus thuringiensis spore and crystal protein to resistant
diamondback moth (Plutella xylostella). Appl. Environ. Microbiol. 62, 564–569
(1996).
52. Zhang, H. et al. Early warning of cotton bollworm resistance associated with intensive
planting of Bt cotton in China. PLoS ONE 6, e22874 (2011).
53. Downes, S., Parker, T. & Mahon, R. Incipient resistance of Helicoverpa punctigera
to the Cry2Ab Bt toxin in Bollgard II cotton. PLoS ONE 5, e12567 (2010).
54.Wan, P. et al. Increased frequency of pink bollworm resistance to Bt toxin Cry1Ac.
PLoS ONE 7, e29975 (2012).
55. Alcantara, E., Estrada, A., Alpuerto, V. & Head, G. Monitoring Cry1Ab susceptibility
in Asian corn borer (Lepidoptera: Crambidae) on Bt corn in the Philippines. Crop
Prot. 30, 554–559 (2011).
56. Huang, F. et al. Extended monitoring of resistance to Bacillus thuringiensis Cry1Ab
maize in Diatraea saccharalis (Lepidoptera: Crambidae). GM Crops Food 3, 245–254
(2012).
57. Downes, S. & Mahon, R. Evolution, ecology and management of resistance in
Helicoverpa spp. to Bt cotton in Australia. J. Invertebr. Pathol. 110, 281–286
(2012).
58. Downes, S. & Mahon, R. Successes and challenges of managing resistance in
Helicoverpa armigera to Bt cotton in Australia. GM Crops Food 3, 228–234 (2012).
59. Ali, M.I., Luttrell, R.G. & Young, S.Y. Susceptibilities of Helicoverpa zea and Heliothis
virescens (Lepidoptera: Noctuidae) populations to Cry1Ac insecticidal protein. J.
Econ. Entomol. 99, 164–175 (2006).
60. Luttrell, R.G., Wan, L. & Knighten, K. Variation in susceptibility of Noctuid
(Lepidoptera) larvae attacking cotton and soybean to purified endotoxin proteins
and commercial formulations of Bacillus thuringiensis. J. Econ. Entomol. 92, 21–32
(1999).
61. McCaffery, A.R. Resistance to insecticides in heliothine Lepidoptera: a global view.
Philos. Trans. R. Soc. Lond., B 353, 1735–1750 (1998).
62. Arthropod Pesticide Resistance Database. Michigan State University. <http://www.
pesticideresistance.com/>
63. Tabashnik, B.E. & Gould, F. Delaying corn rootworm resistance to Bt corn. J. Econ.
Entomol. 105, 767–776 (2012).
64. Crespo, A.L.B. et al. On-plant survival and inheritance of resistance to Cry1Ab toxin
from Bacillus thuringiensis in a field-derived strain of European corn borer, Ostrinia
nubilalis. Pest Manag. Sci. 65, 1071–1081 (2009).
65. Bird, L.J. & Akhurst, R.J. Relative fitness of Cry1A-resistant and -susceptible
Helicoverpa armigera (Lepidoptera, Noctuidae) on conventional and transgenic cotton. J. Econ. Entomol. 97, 1699–1709 (2004).
66. Bird, L.J. & Akhurst, R.J. The fitness of Cry1A-resistant and -susceptible Helicoverpa
armigera (Lepidoptera, Noctuidae) on transgenic cotton with reduced levels of
Cry1Ac. J. Econ. Entomol. 98, 1311–1319 (2005).
67. United States Department of Agriculture Economic Research Service. Adoption of
Genetically Engineered Crops in the US <http://www.ers.usda.gov/data-products/
adoption-of-genetically-engineered-crops-in-the-us.aspx> (2012).
68. Gould, F. et al. Initial frequency of alleles for resistance to Bacillus thuringiensis
toxins in field populations of Heliothis virescens. Proc. Natl. Acad. Sci. USA 94,
3519–3523 (1997).
69. Blanco, C.A. et al. Bacillus thuringiensis Cry1Ac resistance frequency in tobacco
budworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 102, 381–387 (2009).
70. Tabashnik, B.E. et al. DNA screening reveals pink bollworm resistance to Bt cotton
remains rare after a decade of exposure. J. Econ. Entomol. 99, 1525–1530 (2006).
71. Tabashnik, B.E., Dennehy, T.J. & Carrière, Y. Delayed resistance to transgenic cotton
in pink bollworm. Proc. Natl. Acad. Sci. USA 102, 15389–15393 (2005).
72. USDA Agricultural Marketing Service. Cotton Varieties Planted 2012 Crop. <http://
usda.mannlib.cornell.edu/usda/ams/CNAVAR.pdf>
73. Kruger, M.J., Van Rensburg, J.B.J. & Van den Berg, J. Perspective on the development of stem borer resistance to Bt maize and refuge compliance at the Vaalharts
irrigation scheme in South Africa. Crop Prot. 28, 684–689 (2009).
74. Kruger, M.J., Van Rensburg, J.B.J. & Van den Berg, J. Transgenic Bt maize: farmers’
perceptions, refuge compliance and reports of stem borer resistance in South Africa.
J. Appl. Entomol. 136, 38–50 (2012).
75. Storer, N.P. et al. Discovery and characterization of field resistance to Bt maize:
Spodoptera frugiperda (Lepidoptera: Noctuidae) in Puerto Rico. J. Econ. Entomol.
103, 1031–1038 (2010).
76. Storer, N.P., Kubiszak, M.E., King, J.E., Thompson, G. D. & Santos A. C. Status
of resistance to Bt maize in Spodoptera frugiperda: lessons from Puerto Rico. J.
Invertebr. Pathol. 110, 294–300 (2012).
520
77. Carrière, Y. et al. Long-term evaluation of compliance with refuge requirements for
Bt cotton. Pest Manag. Sci. 61, 327–330 (2005).
78. Tabashnik, B.E. et al. Sustained susceptibility of pink bollworm to Bt cotton in the
United States. GM Crops Food 3, 194–200 (2012).
79. Stone, G.D. Biotechnology and the political ecology of information in India. Hum.
Organ. 63, 127–140 (2004).
80. Choudhary, B. & Gaur, K. Bt Cotton in India: A Country Profile. ISAAA Series of
Biotech Crop Profiles (ISAAA, Ithaca, NY, 2010).
81. Herring, R.J. Stealth seeds: Bioproperty, biosafety, biopolitics. J. Dev. Stud. 43,
130–157 (2007).
82. US Environmental Protection Agency. Biopesticides registration action document—
Bacillus thuringiensis plant-incorporated protectants) <http://www.epa.gov/pesticides/biopesticides/pips/bt_brad.htm> (2001).
83. Cotton CRC Extension Team. Cotton pest management guide 2009–10. <http://www.
dpi.nsw.gov.au/__data/assets/pdf_file/0004/154768/cotton-pest-management-guide.
pdf> (2009).
84. US Environmental Protection Agency. Pesticide News Story: EPA Approves Natural
Refuge for Insect Resistance Management in Bollgard II Cotton <http://www.epa.
gov/oppfead1/cb/csb_page/updates/2007/bollgard-cotton.htm> (2007).
85. Mahon, R.J., Downes, S.J. & James, B. Vip3A resistance alleles exist at high levels
in Australian targets before release of cotton expressing this toxin. PLoS ONE 7,
e39192 (2012).
86. Tabashnik, B.E. et al. Efficacy of genetically modified Bt toxins against insects
with different genetic mechanisms of resistance. Nat. Biotechnol. 29, 1128–1131
(2011).
87. Baum, J.A. et al. Control of coleopteran insect pests through RNA interference. Nat.
Biotechnol. 25, 1322–1326 (2007).
88. Mao, Y.-B. et al. Silencing a cotton bollworm P450 monooxygenase gene by plantmediated RNAi impairs larval tolerance of gossypol. Nat. Biotechnol. 25, 1307–1313
(2007).
89. Huvenne, H. & Smagghe, G. Mechanisms of dsRNA uptake in insects and potential
of RNAi for pest control: a review. J. Insect Physiol. 56, 227–235 (2010).
90. Sainsbury, F., Benchabane, M., Goulet, M.-C. & Michaud, D. Multimodal protein
constructs for herbivore insect control. Toxins 4, 455–475 (2012).
91. Lalitha, N., Ramaswami, B. & Viswanathan, P.K. India’s experience with Bt cotton: case studies from Gujarat and Maharashtra. in Biotechnology and Agricultural
Development: Transgenic Cotton, Rural Institutions and Resource-Poor Farmers (ed.
Tripp, R.) 135–167 (Routledge, New York, 2009).
92. Showalter, A.M., Heuberger, S., Tabashnik, B.E. & Carrière, Y. A primer for the
use of insecticidal transgenic cotton in developing countries. J. Insect Sci. 9,
22 (2009).
93. Ali, M.I. & Luttrell, R.G. Susceptibility of bollworm and tobacco budworm (Noctuidae)
to Cry2Ab2 insecticidal protein. J. Econ. Entomol. 100, 921–931 (2007).
94. Jackson, R.E., Gould, F., Bradley, J.R. Jr & Van Duyn, J. Genetic variation for resistance to Bacillus thuringiensis toxins in Helicoverpa zea (Lepidoptera: Noctuidae)
in eastern North Carolina. J. Econ. Entomol. 99, 1790–1797 (2006).
95. Center for Environmental Risk Assessment. GM Crop Database. Event name: MON
89O3 <http://cera-gmc.org/index.php?evidcode%5B%5D=MON89034&auDate1=
&auDate2=&action=gm_crop_database&mode=Submit>
96. Tabashnik, B.E. & Johnson, M.W. Evolution of pesticide resistance in natural enemies.
in Handbook of Biological Control: Principles and Applications (eds. Fisher, T.W. &
Bellows, T.S.) 673–689 (Academic Press, San Diego, 1999).
97. Dhurua, S. & Gujar, G.T. Field-evolved resistance to Bt toxin Cry1Ac in the pink bollworm, Pectinophora gossypiella (Saunders) (Lepidoptera: Gelechiidae), from India.
Pest Manag. Sci. (2011).
98. Bagla, P. Hardy cotton-munching pests are latest blow to GM crops. Science 327,
1439 (2010).
99. Moar, W. et al. Field-evolved resistance to Bt toxins. Nat. Biotechnol. 26, 1072–1074
(2008).
100. United States Department of Agriculture National Agricultural Statistics Service.
Acreage <http://usda.mannlib.cornell.edu/usda/nass/Acre/2000s/2009/Acre-06-302009.pdf> (2009).
101.Van Rensburg, J.B.J. First report of field resistance by stem borer, Busseola fusca
(Fuller) to Bt-transgenic maize. S. African J. Plant Soil 24, 147–151 (2007).
102. Tabashnik, B.E. & Carrière, Y. Resistance to transgenic crops and pest outbreaks.
in Insect Outbreaks Revisited (eds. Barbosa, P., Letourneau, D.K. & Agrawal, A.A.)
341–354 (Wiley-Blackwell, Chichester, UK, 2012)
103. Blanco, C.A. et al. Susceptibility of isofamilies of Spodoptera frugiperda (Lepidoptera:
Noctuidae) to Cry1Ac and Cry1Fa proteins of Bacillus thuringiensis. Southwest.
Entomologist 35, 409–415 (2010).
104.Pedra, J.H.F., McIntyre, L.M., Scharf, M.E. & Pittendrigh, B.R. Genome-wide
transcription profile of field- and laboratory-selected dichlorodiphenyltrichoroethane (DDT)-resistant Drosophila. Proc. Natl. Acad. Sci. USA 101, 7034–7039
(2004).
105. Brent, K.J. Detection and monitoring of resistant forms: an overview. in Pesticide
Resistance: Strategies and Tactics for Management (National Research Council)
298–312 (National Academy Press, Washington D.C., 1986).
106.Whalon, M., Mota-Sanchez, D. & Hollingworth, R.M. Global Pesticide Resistance in
Arthropods (CABI International, Wallingford, UK, 2008).
107. Kruger, M.J., Van Rensburg, J.B.J. & Van den Berg, J. Resistance to Bt maize in
Busseola fusca (Lepidoptera: Noctuidae) from Vaalharts, South Africa. Environ.
Entomol. 40, 477–483 (2011).
108. US Environmental Protection Agency. Current & Previously Registered Section 3
PIP Registrations. <http://www.epa.gov/pesticides/biopesticides/pips/pip_list.htm>
volume 31 number 6 JUNE 2013 nature biotechnology
re v ie w
in Proceedings of the 2004 Beltwide Cotton Conferences, San Antonio, TX,
January 5–9, 2004 (ed. Richter, D.A.) 1373–1383 (National Cotton Council of
America, Memphis, TN, 2004).
118. Luttrell, R.G. & Jackson, R.E. Helicoverpa zea and Bt cotton in the United States.
GM Crops Food 3, 213–227 (2012).
119. Luttrell, R.G. & Ali, M.I. Exploring selection for Bt resistance in heliothines: results of
laboratory and field studies. in Proceedings of the 2007 Beltwide Cotton Conferences,
New Orleans, LA, January 9–12, 2007 (eds. Boyd, S., Huffman, M., Richter, D. &
Robertson, B.) 1073–1086 (National Cotton Council of America, Memphis, TN,
2007).
120. Jackson, R.E., Catchot, A., Gore, J. & Stewart, S.D. Increased survival of bollworms
on Bollgard II cotton compared to lab-based colony. in Proceedings of the 2011
Beltwide Cotton Conferences, Atlanta, GA, January 4–7, 2011 (eds. Boyd, S.,
Huffman, M. & Robertson, B.) 893–894 (National Cotton Council of America,
Memphis, TN; 2011).
121.Williams, M.R. Cotton insect loss estimate. in Proceedings of the 2012 Beltwide
Cotton Conferences, Orlando, FL, January 3–6, 2012 (eds. Boyd, S., Huffman, M. &
Robertson, B.) 1001–1012 (National Cotton Council of America, Memphis, TN, 2012).
122. Jackson, R.E., Bradley, J.R. Jr., Van Duyn, J.W. & Gould, F. Comparative production
of Helicoverpa zea (Lepidoptera: Noctuidae) from transgenic cotton expressing either
one or two Bacillus thuringiensis proteins with and without insecticide oversprays.
J. Econ. Entomol. 97, 1719–1725 (2004).
npg
© 2013 Nature America, Inc. All rights reserved.
(2011).
109. Monsanto Biotechnology Trait Acreage. Fiscal Years 1996 to 2009. <http://www.
monsanto.com/investors/documents/2009/q4_biotech_acres.pdf> Updated:
October 7, 2009.
110. Gassmann, A.J., Petzold-Maxwell, J.L., Keweshan, R.S. & Dunbar, M.W. Field-evolved
resistance to Bt maize by western corn rootworm. PLoS ONE 6, e22629 (2011).
111. Gassmann, A.J., Petzold-Maxwell, J.L., Keweshan, R.S. & Dunbar, M.W. Western
corn rootworm and Bt maize: challenges of pest resistance in the field. GM Crops
Food 3, 235–244 (2012).
112.Gassmann, A.J. Field-evolved resistance to Bt maize by western corn rootworm:
predictions from the laboratory and effects in the field. J. Invertebr. Pathol. 110,
287–293 (2012).
113.Monsanto. Cotton in India. <http://www.monsanto.com/newsviews/Pages/india-pinkbollworm.aspx> (2010).
114. Genetic Engineering Approval Committee. Decisions taken in the 100th Meeting of
the Genetic Engineering Approval Committee (GEAC) held on 12.5.2010. <http://
www.envfor.nic.in/divisions/csurv/geac/decision-may-100.pdf> (2010).
115. Tabashnik, B.E., Gassmann, A.J., Crowder, D.W. & Carrière, Y. Reply to Field-evolved
resistance to Bt toxins. Nat. Biotechnol. 26, 1074–1076 (2008).
116. Tabashnik, B.E. & Carrière, Y. Field-evolved resistance to Bt cotton: Helicoverpa zea
in the US and pink bollworm in India. Southwest. Entomologist 35, 417–424 (2010).
117. Luttrell, R.G. et al. Resistance to Bt in Arkansas populations of cotton bollworm.
nature biotechnology volume 31 number 6 JUNE 2013
521